587,030 active members*
3,774 visitors online*
Register for free
Login
Results 1 to 7 of 7
  1. #1
    Join Date
    Dec 2006
    Posts
    947

    Real World Load Test, Help.

    So I've got my BP clone CNC retrofit up and running on the X and Y. I'm getting about .0005" backlash from my cheap angular bearings, which is solved perfectly by Mach backlash comp.

    The question is, is there a real world load test to see if the backlash exaserbates as it is milling something like steel? I mean the table pulls me without loosing steps or position but that's once the backlash is taken up and the balls in the nut are already against that side of the nut so putting weight on table as it moves doesn't tell me much. I'm just wondering if backlash will stay the same. I think it will.

    I know not the right forum for the next question but most of you guys know about mach. I enabled backlash comp in mach, do I need to change the number of lines for the look ahead? Currently it's at 20.

  2. #2
    Join Date
    Jul 2005
    Posts
    12177
    One real world test for backlash is to interpolate a circle at faster and faster feeds and see if it remains a circle.

    Alternatively do a square both inside and outside and check how non-square it is at faster feeds.
    An open mind is a virtue...so long as all the common sense has not leaked out.

  3. #3
    Join Date
    Dec 2006
    Posts
    947
    What am I going to do with a whole bunch of circles and squares? LOL.

    Thanks, makes sense. I've heard the term but what do you mean by interpolate? Isn't a circle a circle or is it the way it's machined?

  4. #4
    Join Date
    Jul 2005
    Posts
    12177
    Quote Originally Posted by Cartierusm View Post
    .....I've heard the term but what do you mean by interpolate? Isn't a circle a circle or is it the way it's machined?
    It is early in the morning and I am having my first cup of horrible tasting coffee (my own brew), my machines are on their Warm-up cycle, so I decided to give you the long answers to your questions.

    Circular Interpolation is the way a circle is machined on a milling machine.

    You ask; "Isn't a circle a circle(?)". The definition I recall from math class is; ' A circle is the path traced out by a point moving in a plane at a constant distance from a fixed point in that plane.'. You probably learnt the same thing maybe just worder differently.

    Lathes make circles really easily; the part rotates about a fixed point and the tool traces out the circle. If you were sitting on the rotating part you would see the tool going around you exactly like the definition.

    On a (three axis) milling machine it is more complicated; the part can only move in a straight line along the X or Y axis so the motion needed to create a circular path has to be obtained from these straight moves. I will not bring in a full blown definition of interpolate because they are even more long winded than me, but a simple definition is; 'fill in the gaps'.

    A milling machine creates a circle by doing a series of tiny straight moves on the X and Y axes; it fills the circular path between all the points on a circle with a whole lot of really tiny straight moves. In other words it interpolates a circle from straight moves.

    Straight moves on a milling also have to be interpolated unless they are exactly along either the X or Y axis; any diagonal move has to be a series of X, Y moves that are very small.

    So that is why it is called Circular Interpolation and Linear Interpolation.

    And all this is correct for Stepper driven machines because Steppers move in steps. With Servos the speed can be very accurately controlled so a linear move need not be a series of steps it can be a rtrue straight line created by running the X and Y servos at the appropriate speed.

    Even circles can be generated by constantly varying the servo speed on the two axes so if you want to be really picky you could say servo driven machines do linear cuts and circular cuts because they don't need to interpolate.

    However backlash still creates distortions in the circular path because no matter how the axes are driven they have to reverse direction to get back to the start point.

    There you have it; one coffee cup full of typing and my machines are probably all ready for full speed operation.
    An open mind is a virtue...so long as all the common sense has not leaked out.

  5. #5
    Join Date
    Apr 2005
    Posts
    1194
    Running a bullbar is the way to compensate for this as far as circles.

    You can also get very close setting backlash like this. You use a indicator and establish a zero ont he dial against a block on the table then you jog or rapid back about .030-.100 and rapid back up to the block. Note the change in the reading that then you can use the change as your backlash setting. Do this in X/y/z and you should be good to go.
    We have had good luck with our Fadals milling mostly soft steel and aluminum up to 5 axis. We are always looking for spare parts If you have a broken down Fadal give a shout.

  6. #6
    Join Date
    Dec 2006
    Posts
    947
    So when you say interpolate a circle, why not just say cut a circle? Should I be making my G code in a special way for my steppers?

  7. #7
    Join Date
    Jul 2005
    Posts
    12177
    Quote Originally Posted by Cartierusm View Post
    So when you say interpolate a circle, why not just say cut a circle? Should I be making my G code in a special way for my steppers?
    I guess just saying cut a circle is fine. The G-code commands for cutting circles are G02 for clockwise and G03 for counterclockwise and these are called Circular Interpolation commands.

    You don't do anything special in the G-code, the command:

    G03 I0. J-.5

    will tell most machines to cut a counterclockwise circle around a center point located 0. distance away on the X axis and -0.5 distance away on the Y axis.

    There is some variation in how machines interpret the I and J values because in some cases I0. J-.5 means the center point is located at this absolut coordinate location.
    An open mind is a virtue...so long as all the common sense has not leaked out.

Similar Threads

  1. How evil is a P-Channel MOSFET in the real world?
    By CrazyIvan in forum Open Source Controller Boards
    Replies: 0
    Last Post: 03-25-2009, 07:48 AM
  2. use of G10 in the real world.
    By bookwurm99 in forum G-Code Programing
    Replies: 20
    Last Post: 12-10-2007, 09:43 PM
  3. 3D modelling in the real world
    By ksealey in forum WoodWorking Topics
    Replies: 2
    Last Post: 11-30-2007, 08:06 AM
  4. Mill Drill Real World CNC Feed Rate
    By wmgeorge in forum Benchtop Machines
    Replies: 1
    Last Post: 09-28-2005, 08:39 PM
  5. Any real world experience with Camsoft & Steppers
    By mbam in forum CamSoft Products
    Replies: 20
    Last Post: 06-23-2004, 04:59 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •